Vascular endothelial dysfunction in sickle cell anemia (SCA) leads to acute and chronic organ complications.1 The vasculopathy may be due, in part, to intravascular hemolysis releasing cell-free hemoglobin and heme into the circulation leading to direct oxidative injury, depletion of nitric oxide, and upregulation of inflammatory and immune response pathways.1

Thrombomodulin (THBD) is a transmembrane protein that functions at the luminal surface of vascular endothelial cells to bind thrombin and inhibit its interaction with fibrinogen, augment protein C activation, and down-regulate complement activation.2 Oxidative stress can reduce THBD function by inducing extracellular proteolytic cleavage of THBD from the endothelial surface or down-regulating THBD expression in endothelial cells.2 Reduced endothelial THBD function has been implicated in atypical hemolytic uremic syndrome, disseminated intravascular coagulation, and pre-eclampsia.2 Increased THBD in circulation, implying reduced membrane THBD function, predicts clinical severity in these microangiopathic disorders. The role of THBD function and the clinical significance of increased THBD in the plasma in SCA is unclear.

We investigated whether cell-free hemoglobin reduces THBD function and whether impaired THBD function is implicated in the pathophysiology of SCA vasculopathy by studying endothelial cells and a prospective cohort of patients with SCA.

Endothelial cells (EA.hy926, ATTC® CRL-2922™; Manassas, VA) were exposed to incremental doses of cell-free hemoglobin (Sigma-Aldrich; St. Louis, MO). We focused our experiments on the 2- and 6-h time points because we observed increased cell expansion and crowding in the control conditions versus a reduction in viable cell numbers in the cell-free hemoglobin-exposed endothelial cells at 24 h of incubation (Figure S1). The concentrations of THBD in the supernatant and in the plasma of SCA patients were determined by enzyme-linked immunosorbent assay (R&D Systems; Minneapolis, MN). THBD activity on the endothelial cell surface was assessed by cleavage of chromogenic substrate for activated protein C in the presence of thrombin.3 For the immunofluorescence studies, endothelial cells were grown in Nunc™ Lab-Tek™ CC2™ chamber slides (ThermoScientific; Waltham, MA) and exposed to cell-free hemoglobin for 6 h. Cells were then fixed with 1% paraformaldehyde solution (ThermoFisher Scientific; Waltham, MA), stained with human THBD (1009) mouse monoclonal antibody (Cell Marque; Rocklin, CA) and goat anti-mouse Alexa Fluor Plus 488 secondary antibody (ThermoFisher Scientific; Waltham, MA), and then treated with an auto-fluorescence quenching reagent (Vector Labs; Burlingame, CA). The cell nuclei were stained with 4′,6-diamidino-2-phenylindole (DAPI, ThermoFisher Scientific; Waltham, MA) and all washes were done with PBS. The endothelial cells were mounted with the Vectashield® Vibrance™ antifade mounting medium (Vector Labs; Burlingame, CA), followed by fluorescent quantification using ImageJ and images being taken at 60X magnification (Olympus BX51/IX70; Tokyo, Japan).

Ninety SCA patients (88 Hb SS, 2 Hb Sβ0-thalassemia) recruited into a longitudinal kidney study between March 2013 and November 2017 were monitored for acute multiorgan failure.4 The protocol was approved by the UIC Institutional Review Board and patients provided written informed consent prior to recruitment and biosample collection during an outpatient clinic visit. Hemoglobinuria was defined as urine dipstick positive for blood and < 2 red blood cells/high power field by microscopy. Acute kidney injury (AKI) was defined and staged according to the Kidney Disease Improving Global Guidelines as follows: Stage 1 = serum creatinine rise 1.5–1.9 times baseline, Stage 2 = serum creatinine rise 2.0–2.9 times baseline, Stage 3 = serum creatinine rise ≥ 3 times baseline, ≥ 4.0 mg/dl, or requiring hemodialysis.5 The estimated glomerular filtration rate (eGFR) was calculated using the Chronic Kidney Disease Epidemiology Collaboration formula.6 Between September 2020 and May 2021, we recruited and collected blood samples from an additional 44 SCA patients within 24 h of hospitalization for a vaso-occlusive crisis (VOC).

Comparisons of plasma THBD concentrations in SCA patients by hemoglobinuria status and by multiorgan failure status during hospitalization for a VOC were conducted using the Mann–Whitney test. The correlation of plasma THBD concentrations obtained at baseline versus at a repeat outpatient time point was assessed using linear regression analysis. The association of the highest plasma THBD concentration by quartile, defined as THBD > 4.23 μg/ml, with multiorgan failure was examined using the logrank method to compare Kaplan–Meier survival curves and Cox proportional hazards models, adjusting for age, sex, hydroxyurea use, and eGFR. Multiorgan failure events were censored at the date of last contact. Systat 13 (Systat Software Corporation; Chicago, IL) was used for analyses.

We first examined whether cell-free hemoglobin is associated with the release of THBD into the supernatant and reduced endothelial THBD activity. Increased THBD was observed in the supernatant of endothelial cells at 6 h after exposure to incremental doses of cell-free hemoglobin (Figure 1A). Consistent with increased loss of THBD from the endothelium into the supernatant, a reduction in THBD activity and decreased THBD by immunofluorescence was observed on the endothelial cell surface at 6 h after exposure to cell-free hemoglobin (Figure 1B,C).

(A) Increased concentration of THBD in the supernatant and (B) declining THBD activity on the endothelial cell surface after exposure to incremental doses of cell-free hemoglobin. Figures A and B represent combined results from five independent experiments; * signifies p < .05 and ** signifies p < .01. (C) Reduced THBD by fluorescent intensity on the endothelial cell surface after 6 h of exposure to incremental doses of cell-free hemoglobin. (D) Higher THBD concentrations in SCA patients with versus without hemoglobinuria. (E) Plasma concentrations of THBD were stable among SCA patients on repeat sampling (n = 23). (F) Plasma THBD concentrations in the highest quartile were associated with a greater risk for developing a multiorgan failure event (10/15 vs. 9/56, respectively). (G) Plasma THBD concentrations were significantly higher in SCA patients that developed multiorgan failure during a VOC compared to levels from SCA patients during a clinic visit or during hospitalization for VOC without multiorgan failure. VOC, vaso-occlusive crisis; MOF, multiorgan failure

In a longitudinal cohort of 90 SCA patients, we examined whether plasma THBD concentrations are associated with exposure to cell-free hemoglobin and could serve as a biomarker for the risk of multiorgan failure syndrome and survival, similar to what is observed in other microangiopathic disorders.2 The median age of this cohort was 35 years (interquartile range [IQR], 28–43 years), 48% were female, and 46% were on hydroxyurea at the time of enrolment. We observed higher plasma THBD concentrations in SCA patients with versus without hemoglobinuria, a biomarker of intravascular hemolysis that represents cell-free hemoglobin that has been filtered through the glomerulus from the circulating blood (Figure 1D). We were able to resample 23 SCA patients on routine follow-up. With a median time interval of 2.1 years (IQR, 1.8–2.4 years), the concentration of THBD in the follow-up plasma samples remained consistent compared to their respective baseline values (Figure 1E).

With a median follow-up of 6.9 years (IQR, 3.2–7.6 years), we observed 24 multiorgan failure events in 19 SCA patients (22 during hospitalization for vaso-occlusive crises and 2 in the setting of severe pre-eclampsia). Pulmonary, renal, neurologic, cardiovascular, and hepatic organ systems were commonly affected (Figure S2; Table S1). Twenty-one percent (5/24) of the multiorgan failure events led to death. After adjusting for age, sex, hydroxyurea use, and eGFR, baseline plasma THBD concentration in the highest quartile was an independent predictor for developing a multiorgan failure event (HR 5.2, 95% CI: 1.3–21.4; p = .02) on longitudinal follow-up (Figure 1F).

Between September 2020 and May 2021, we collected blood from 44 SCA patients within 24 h of hospitalization for a VOC from all patients who provided consent. Plasma THBD concentrations were significantly higher in the SCA patients hospitalized for a VOC complicated by multiorgan failure (n = 6) compared to levels from SCA patients hospitalized for a VOC but did not develop multiorgan failure (n = 38) (Figure 1G).

Increased THBD in circulation is a biomarker of endothelial cell injury that correlates with disease severity in systemic vasculitis disorders, such as systemic lupus erythematosus or Wegener's graulomatosis, and in microangiopathic hemolytic anemias, such as atypical hemolytic uremic syndrome, thrombotic thrombocytopenic purpura, and disseminated intravascular coagulation.2 Our in vitro and SCA patient data suggests that cell-free hemoglobin exposure may mediate loss of THBD from the endothelium and reduce the protective effects of THBD on vascular endothelium. Furthermore, increased plasma concentrations of THBD predicts a 5-fold greater risk of multiorgan failure in SCA patients on longitudinal follow-up. Limitations of our study include that we did not have a sufficient sample size to assess the relationship between THBD risk variants implicated in SCA-related thrombosis7 and THBD levels. Furthermore, we did not sequentially measure concentrations of cell-free hemoglobin or heme and THBD during hospitalization to determine whether these levels increased prior or subsequent to the development of multiorgan failure. Future studies of THBD in SCA vasculopathy in larger cohorts with sequential biosampling may guide our understanding and the development of interventions for the catastrophic multiorgan failure syndrome.

CONFLICTS OF INTEREST

The authors declare no relevant competing financial interests.

AUTHOR CONTRIBUTIONS

Maria Armila Ruiz, Binal N. Shah and Santosh L. Saraf designed and performed research, analyzed the data, and wrote the paper. Guohui Ren, David Shuey, Richard D. Minshall, and Victor R. Gordeuk designed and performed research and wrote the paper.